Re: [PATCH] reformat mtrr.c to conform to CodingStyle

[David Weinehall]

On Tue, Nov 13, 2001 at 12:10:22PM -0700, Andreas Dilger wrote:
> On Nov 12, 2001  23:25 -0500, Benjamin LaHaise wrote:
> > Please incorporate this patch to make mtrr.c conform to the standards set 
> > forth in Documentation/CodingStyle which make it much more appealing to 
> > the eyes.
> >
> >  /*  Put the processor into a state where MTRRs can be safely set  */
> > -static void set_mtrr_prepare (struct set_mtrr_context *ctxt)
> > +static void
> > +set_mtrr_prepare(struct set_mtrr_context *ctxt)
> >  {
> 
> Is that actually CodingStyle?  Don't see it much in the kernel code...
> The much more common (AFAICS) style to split long function definitions is
> 
> static void foo_long_function(struct long_struct name *foo, struct bar *bar,
>                               int val, long *err)
> 
> The only reason (AFAICS) for putting the return type on a separate
> line is the (ancient) ansi2knr script, which just throws the return
> types away for pre-ANSI compilers.  Given that the kernel code doesn't
> even conform to ANSI-C, there is no hope in hell of it compiling with
> a pre-ANSI compiler.

grep:ing for the function-name using ^fname is a common use. Easily
solvable anyway, but...

I don't think Lindent does everything 100% correct; at least its
formatting of switch/case does look a little fishy:

	switch (option) {
	case 1: 
		/* blaha */

That feels kind of odd compared to the rest of the codingstyle.

Comments?!

/David Weinehall
  _                                                                 _
 // David Weinehall <[e-mail]> /> Northern lights wander      \\
//  Maintainer of the v2.0 kernel   //  Dance across the winter sky //
\>  http://www.acc.umu.se/~tao/    </   Full colour fire           </